Re: Random Regression - maternal effects

From: Arthur Gilmour <arthur.gilmour_at_CARGOVALE.COM.AU>
Date: Mon, 6 Apr 2009 09:54:53 +1000

Dear Boyd,

ASReml returns the Wald F statistic for the whole leg(age,p) term
but you would like to know whether the highest order term is
significant. I.e. is leg(age,3) significantly better than
leg(age,2).

Now the first 3 columns of the design matrix for leg(age,3) are
identical to the 3 columns of the design matrix for leg(age,2).

So, look at the .sln file and find the 4 coefficents for the leg(age,3)
term. Consider the 4th coefficient and construct a t statistic as its
ratio to its standard error.

You can convert that to an F statistic by squaring it.

This F value is the F value for adding this effect last to the whole
model. So, given you also know the denominatior degrees of freedom
(assume it is the same for all 4 effects and ASReml can calculate it
using !DDF). You can test whether the highest order term is
significant.

If you decide to drop it, you will need to refit the (reduced) model to
test the next term.

You should note reduce it below the power involved in any interaction
(random) terms.

I trust that helps.

On Sun, 2009-04-05 at 01:55 +0100, bgudex wrote:
> Thanks Arthur.
>
> One other question - I've been asking a few people around here what
> the best way of determining what order polynomial to use in the fixed
> regression component of the random regression model?
>
> It has been suggested that I consider either the F value for
> leg(age,1) but it doesn't appear in my output files or to compare the
> var(e) from different orders of fit. Is there any way of getting an F
> value?
>
> Many thanks
> Boyd
>
>
>
>
> -------------------- m2f --------------------
>
> Sent using Mail2Forum (http://www.mail2forum.com).
>
> Read this topic online here:
> http://www.vsni.co.uk/forum/viewtopic.php?p=476#476
>
> -------------------- m2f --------------------
>
>
>
>
> This message is intended for the addressee named and may contain confidential information. If you are not the intended recipient, please delete it and notify the sender. Views expressed in this message are those of the individual sender, and are not necessarily the views of their organisation.

```--
May the God and Father of the Lord Jesus Christ guide and bless you.
Arthur Gilmour
PS Are you anxious?
http://www.aish.com/spirituality/growth/God_vs._Prozac.asp
Mobile Number +61 427 227 468
Home phone +61 2 6364 3288  Skype: Arthur.Gilmour
http://www.CargoVale.com.au
Wisdom is justified by all her children  Luke 7:35
Travel:
USA  1-11 April
Tas 11-16 May
Brazil Jul27 - Aug 8
UK Aug 10-14
```
Received on Tue Apr 06 2009 - 09:54:53 EST

This webpage is part of the ASReml-l discussion list archives 2004-2010. More information on ASReml can be found at the VSN website. This discussion list is now deprecated - please use the VSN forum for discussion on ASReml. (These online archives were generated using the hypermail package.)